Table 1.

Inclusion and exclusion criteria for each PECOS domain in relation to long-term exposure and health effects to selected air pollutants.

PECOS	Inclusion	Exclusion
Population	•General human population (including sub-groups at risk: children, pregnant women, elderly, or patients with particular conditions), of all ages, developed and developing areas, both urban and rural. No geographical restrictions. •Study population expose to the pollutant of interest via inhalation through ambient air predominantly	•Study population expose to the pollutant of interest in occupational settings or indoor exposure exclusively
Exposure	•Long-term exposure (order of years) to ambient air O3 and NO2 expressed in a concentration unit (ppb and μg/m3 respectively).	•Less than one year of data available
Comparator	•Exposure to per concentration increased unit of the air pollutant of interest in the same population	•Increment for hazard ratio not given
Outcome	•Health outcomes selected in relation to long-term exposure include (ICD 10 codes, version 2016 in brackets): all cause (A00-Z99); respiratory (J00-J99); COPD (J40-47) and ALRI (J12-J18, J20-J28) mortality [Note: Studies vary in selection of codes.]	•Birth outcomes (due to neonatal exposure of pollutant)
Study	• Human epidemiological studies including: o Prospective and retrospective cohort studies •Published (or accepted for publication i.e. in press) journal articles in any language (abstract in English language), conference abstracts and papers, letters, notes, grey literature. •If suitable articles are identified published in languages not known by the SRT, further assistance will be sought after (members of the GDG or external review team from different regions, colleagues, researcher networks, etc)	•Qualitative studies •Case control studies •No adjustment for socio-economic status (individual or area) •Studies where no original data were analysed •Reviews and methodological papers •Non-human studies (in vivo, in vitro, other) •Insufficient information given to standardise hazard ratio and precision (standard error or confidence interval)

SRT: Systematic review team